Figure 1.
Folding of the membrane sensor region of Spo20 onto anionic liposomes.
(A) Domain organization of Spo20. The sequence of the membrane-binding region is indicated. The helical wheel representation, which was drawn using heliquest (http://heliquest.ipmc.cnrs.fr/) [28], illustrates the amphipathic character of its central part. Hydrophobic residues are shown in yellow, arginines and lysines in dark blue, histidines in light blue, glutamine and asparagine in pink and serine in purple. (B) Far-UV CD spectra of the Spo20 peptide (aa 56–79; 30 µM) in solution (dark blue curve) or with 3 mM of small liposomes (Rh = 15±4 nm) containing increasing % of either PA, PS, or PIP2 (black to cyan curves). The remaining lipids in the liposome were (in mol %) PE (25), and PC (the concentration of which varied from 75 to 45% depending on the concentration of anionic lipid). MRE, mean residue ellipticity. (C) Left: helicity at 222 nm as a function of the percentage of PA, PS or PIP2 in the liposomes and as determined from the spectra shown in panel B. Right: helicity at 222 nm as a function of the amount of negative charges carried by PA, PS or PIP2 in the liposomes.
Figure 2.
A bioprobe strategy to assess the membrane binding properties of the Spo20 sensor region.
(A) Probe design. The probes were constructed according to the design of the N-terminal region of the Golgin GMAP-210. The ALPS motif of GMAP-210 was replaced by the Spo20 membrane region. In the case of the Spo20 probe, the coiled coil of GMAP was shortened by including a stop codon at position K202 and its endogeneous cysteines were mutagenized. The obtained construction, Spo20-GCC, is more soluble and was used for flotation and NBD-fluorescence experiments. (B) Membrane partitioning of the NBD labeled forms of Spo20-GCC or ALPS-GCC (0.3 µM) in the presence of liposomes of defined size as obtained by extrusion (0.35 mM lipids). The liposomes contained (mol %): PC (40), PE (30), cholesterol (25) and PA (5). (C) Flotation experiments. Spo20-GCC but not ALPS-GCC, bound to PA containing liposomes. Spo20-GCC or ALPS-GCC (0.75 µM) was incubated without liposomes or with liposomes (0.75 mM lipids) containing (mol %): PE (25), cholesterol (25), PS (15) and increasing amount of PA (0 to 15). The remaining lipid was PC. Liposomes were extruded through 0.2 µm filters.
Figure 3.
Subcellular localization of coiled-coil bioprobes containing ALPS or the Spo20 membrane sensor region.
(A) GCC-GFP, ALPS-GCC-GFP or Spo20-GCC-GFP in RPE1 cells. (B) Coexpression of monomeric Spo20-GFP and Spo20-ACC1-mCherry. The two constructs labeled the same membrane structures (plasma membrane ruffles), but Spo20-GFP also invades the nucleus and is more cytosolic than Spo20-ACC1-mCherry. (C) Coexpression of Spo20-ACC1-GFP and ACC1-mCherry or ACC1-GFP and ACC1-mCherry. Due to dimerisation, three coiled-coil probes should form with different color properties and various numbers of Spo20 membrane sensor regions. The moderate labeling of plasma membrane ruffles by mCherry (arrows) indicates that the heterodimer Spo20-ACC1-GFP/ACC1-mCherry, which contains one Spo20 motif, binds membranes. Scale bars, 10 µm.
Figure 4.
Membrane binding properties of the Spo20 bioprobe.
(A) Spo20-GCC (0.75 µM) was incubated with or without PC/PE/Cholesterol liposomes (0.75 mM) containing (mol %) PE (25), cholesterol (25) and supplemented with PA (15), PS (30), or PIP2 (5). The remaining lipid was PC. Bound proteins were recovered by flotation and analyzed by SDS-polyacrylamide gel electrophoresis using Sypro orange staining. (B, C) Binding of Spo20-GCC to liposomes containing (mol %) PE (25), cholesterol (25) and increasing amounts of PA (circles) or PS (triangles) as indicated. The remaining lipid was PC. A dose response curve for PA with liposomes containing 25 mol % PE, 25 mol % cholesterol and 15 mol % PS is also shown (squares). Membrane partitioning was assessed either by the NBD fluorescence assay (B) or by the liposome flotation assay (C). The data shown for the NBD assay are from two independent experiments. The horizontal dashed line indicates the fluorescence level of the NBD proteins in solution. The vertical bars show the standard deviation of the NBD fluorescence intensity of Spo20-GCC. For the flotation assay, the data shown are from two or three independent experiments with different preparations of liposomes. An typical SDS gel analysis is shown. The dose response curves are shown either as a function of the mol % of PA or PS (left) or as a function of the total amount of negative charges in the membrane (right). (D) Effect of PE on the membrane partitioning of Spo20-GCC as assessed by NBD fluorescence. The liposomes contained (mol %) PS (15), cholesterol (25), PA (0, triangles; 10, squares) and increasing amounts of PE at the expense of PC. All liposomes were prepared by extrusion through 0.2 µm filters.
Figure 5.
Real-time measurement of PA production by an extract containing PLD activity.
The fluorescence cuvette contained liposomes (0.35 mM lipids; extrusion 0.2 µm) with 25 mol % PE, 25 mol % cholesterol supplemented or not with 5 mol % PIP2. The remaining lipid was PC. Then, [NBD]Spo20-GCC (0.3 µM) and a PLD extract from Streptomyces chromofuscus (25 to 100 U/ml) were sequentially added. Experiments were carried out in triplicates; representative traces are shown. Inset: comparison between PLD and boiled PLD under the same conditions.
Figure 6.
Membrane binding properties of the swap Spo20 mutant.
(A) Helical wheel representations of the Spo20 sequence and of the corresponding swap mutant. (B–C) NBD fluorescence assays comparing the membrane partitioning of the [NBD]Spo20-GCC and [NBD]swapSpo20-GCC bioprobes (0.3 µM) to various liposomes (0.35 mM). In (B), the liposomes contained (mol %) PE (25), cholesterol (25), PS (0 or 15) and increasing amounts PA. The remaining lipid was PC. In (C), the liposomes contained (mol %) PC (25), PE (25), cholesterol (25), PA (10) and PS (15) and the partitioning of the constructs was tested at various pH. Data shown are mean ± S.E of 3 independent experiments. All liposomes were prepared by extrusion through 0.2 µm polycarbonate filters.
Figure 7.
Inverting the Spo20 sequence affects neither its specificity for negatively charged lipids nor its subcellular localization.
(A) Helical wheel representations of the Spo20 sequence and of the corresponding inverted mutant (iSpo20) where the sequence is read from the C- to the N-terminus. (B) Flotation Assays. Binding of iSpo20-GCC to liposomes (0.75 mM lipids; extrusion 0.2 µm) containing (mol %) PE (25), cholesterol (25), PS (0, open symbols; 15, filled symbols) and increasing amounts of PA. The experiment was repeated two or three times with different preparations of liposomes. Data show mean ± S.E of these independent experiments. (C) Confocal microscopy images of RPE1 cells after transfection with Spo20-ACC1-mCherry and iSpo20-ACC2-GFP. xy planes and z projections are shown. Note that the two probes co-localize almost perfectly. Scale bars = 10 µm.
Figure 8.
Effect of PLD2 overexpression on the subcellular localization of coiled-coil probes.
(A) Spo20-GCC-GFP and ALPS-ACC1-mCherry were coexpressed in RPE1 cells. Because the two constructs contain different coiled-coil domains they form homodimeric probes. Spo20-GCC-GFP/Spo20-GCC-GFP stains ruffles at the plasma membranes, whereas ALPS-ACC1-mCherry/ALPS-ACC1-mCherry stains the Golgi apparatus. (B) Coexpression of Spo20-ACC1-GFP and ALPS-ACC1-mCherry. Because the two constructs contain the same coiled-coil, they not only homodimerize but also form the heterodimer Spo20-ACC1-GFP/ALPS-ACC1-mCherry. The dual color of the Golgi apparatus (green+red = yellow) indicates that the heterodimer is directed towards this compartment. (C) Same as in (B) but in cells overexpressing PLD2. Some plasma membrane ruffles where PLD2 localizes appear yellow, whereas the Golgi apparatus is completely red, suggesting that the Spo20-ACC1-GFP/ALPS-ACC1-mCherry heterodimer is redirected from the Golgi apparatus to the plasma membrane upon the action of PLD2. Scale bars = 10 µm.